IATF 7.1.5.1.1 Measurement system analysis (Visual Inspection)

I know what you are saying - I’m rejecting the premise of the statement, not just the statement.

I’m aware of the paper and it’s critiques. The problem with it is even they say that there can be 20 - 30% error rate with simple inspections. Their study wasn’t comprehensive… Hardly a ringing endorsement of the ‘average’ of 80% effective. Besides it’s not the average that matters it’s the standard deviation that will kill you. Or save your butt. And those who typically quote the 80% do do because “they head it somewhere” and “everyone knows it”. They have no clue as to the scope, data or findings in the paper.

We also need to remember that there are a lot of studies with conclusions that have proven to be 1) not supported by the actual data and/or 2) shown to be just coincidental with a confounding factor or 3) completely mis-interpreted (cherry picked from a highly specific constrained situation to a global generalization) Ranging from the Hawthorne effect to the truly stupid 1.5 sigma shift to the abomination that autism is caused by vaccinations.